/[svn]/linuxsampler/trunk/ChangeLog
ViewVC logotype

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 1829 - (show annotations) (download)
Fri Jan 30 19:22:36 2009 UTC (15 years, 1 month ago) by iliev
File size: 25610 byte(s)
* added (experimental) CoreAudio driver

1 Version CVS HEAD (?)
2
3 * packaging changes:
4 - autoconf bugfix: the PKG_CONFIG variable wasn't initialized properly,
5 causing e.g. the libgig test to fail when
6 "./configure --disable-jack-driver" was used
7 (patch by Alexis Ballier)
8 - fixed compilation with gcc 4.3
9 - fixes for building on OS X (thanks to Ebrahim Mayat for testing)
10 - fixed configure so it detects x86_64 (#107)
11 - fixes for building with newer MinGW versions
12 - fix for building with bison 2.4 (#111)
13
14 * general changes:
15 - bugfix: on some POSIX systems instrument editor plugins refused to
16 load as we used a non-portable Linux specific struct field
17 (fixes bug #70, patch by Ronald Baljeu)
18 - fixed endless loop which occured when loading an instrument editor
19 plugin DLL on Windows systems failed
20 - fixed memory leaks that occurred when liblinuxsampler was unloaded
21 - optimized the SynchronizedConfig class so it doesn't wait
22 unnecessarily long after an update
23 - added support for notifying instrument editors on note-on / note-off
24 events (e.g. to highlight the pressed keys on the virtual keyboard
25 of gigedit)
26 - added support for triggering notes by instrument editors (see above)
27 - be verbose on DLL load errors (on Linux)
28 - fixed di-harmonic triangle LFO implementation (this LFO implementation
29 is only used by very few systems, most pick the int math
30 implementation, which is usually faster)
31 - fixes for audio drivers with varying buffer sizes
32 - experimental support for running LinuxSampler as a DSSI, LV2 and
33 VST plugin
34 - notification events for stream/voice count statistics are now sent
35 only when there are actual changes
36 - added memory ordering constraints to improve stability on
37 multi-core and multi-cpu systems
38 - maximum voices and disk streams can now be altered at runtime
39 - fixed CPU feature detection on x86_64 (maybe fixes #108)
40
41 * audio driver:
42 - removed the nonsense audio channel constraint (which was hard coded to
43 max. 100 audio channels) for most audio drivers
44 - JACK audio driver did not offer a device parameter "SAMPLERATE" as
45 opposed to the LSCP specs
46 - bugfix: the SAMPLERATE parameter of some drivers (e.g. JACK)
47 reflected the wrong value
48 - fixed a memory management bug in ASIO driver
49 - Makefile fix: JACK_CFLAGS wasn't used
50 - JACK: use jack_client_open instead of the deprecated
51 jack_client_new
52 - added (experimental) CoreAudio driver
53
54 * MIDI driver:
55 - added JACK MIDI driver
56 - dispatch bank select as ordinary CC as well, the user might seriously
57 want to (mis)use it for some purpose ("fixed" in all current MIDI
58 input drivers: ALSA, CoreMIDI, JACK, MidiShare, MME)
59
60 * instruments database:
61 - avoid time consuming samples scanning when adding instruments
62 to the instruments database
63 - added support for handling lost files in the instruments database
64 - Implemented option for adding instruments in separate directories
65 in the instruments database
66 (patch by Chris Cherrett & Andrew Williams, a bit adjusted)
67
68 * Gigasampler format engine:
69 - fixed a memory leak that could happen when a channel was deleted
70 while notes were playing
71 - made it possible to load an instrument even if the audio thread
72 isn't running
73 - added partial support for the "Controller Triggered" MIDI rule,
74 enough for piano gigs with pedal noise samples
75 - minor fix: only mark FX sends as being modified if really the
76 respective FX send MIDI controller was used
77 - added support for GM global device volume SysEx message
78 (can be selected at compile time to either apply globally to the
79 whole sampler [default] or only to the sampler channels that are
80 connected to the respective MIDI input port the SysEx message arrived
81 on)
82 - bugfix: notes triggered at position 0 in the audio buffer were
83 sometimes wrongly killed in the same buffer, causing no sound to
84 be played
85 - added support for chromatic / drumkit mode Roland GS Sysex message
86 which is usally used to switch a part between chromatic sounds and
87 drumkit sounds (as opposed to their common default setting of MIDI
88 part 10 being a drumkit part and all other ones chromatic parts), in
89 LS however one can switch between the first 16 MIDI instrument maps
90 defined for the sampler
91 - made it possible to create multiple sample channels even if the
92 audio thread isn't running
93
94 * LSCP server:
95 - added new LSCP event "CHANNEL_MIDI" which can be used by frontends to
96 react on MIDI data arriving on certain sampler channels (so far only
97 Note-On and Note-Off events are sent via this LSCP event)
98 - added new LSCP event "DEVICE_MIDI" which can be used by frontends to
99 react on MIDI data arriving on certain MIDI input devices (so far only
100 Note-On and Note-Off events are sent via this LSCP event)
101 - added new LSCP commands: FIND LOST DB_INSTRUMENT_FILES and
102 SET DB_INSTRUMENT FILE_PATH
103 - added new LSCP command: SEND CHANNEL MIDI_DATA which can be used by
104 frontends to send MIDI messages to specific sampler channel
105 - added two additional fields to GET FILE INSTRUMENT INFO command -
106 KEY_BINDINGS and KEYSWITCH_BINDINGS
107 - bugfix: the bank number provided by MIDI_INSTRUMENT_INFO notifications
108 was incorrect
109 - Optimized the retrieval of the MIDI instrument mappings
110 - added new LSCP command "SET VOICES" to globally alter the maximum
111 amount of voices, added new LSCP event "GLOBAL_INFO:VOICES" which will
112 be triggered respectively
113 - added new LSCP command "SET STREAMS" to globally alter the maximum
114 amount of disk streams, added new LSCP event "GLOBAL_INFO:STREAMS"
115 which will be triggered respectively
116 - bugfix: retry if "select" returns EINTR (this fixes a crash when
117 a gigedit file dialog is opened)
118
119 * Bug fixes:
120 - fixed a crash which occurs when removing a sampler channel waiting
121 to start instrument loading after another channel
122 - fixed a crash which occurs when removing a sampler channel with
123 instrument loading in progress (bug #113)
124 - fixed termination caused by uncaught exception when adding MIDI
125 instrument with PERSISTENT load mode
126
127
128 Version 0.5.1 (6 December 2007)
129
130 * packaging changes:
131 - added autoconf checks for pthread library
132 - added autoconf check for pthread bug found on certain NPTL-enabled
133 glibc versions (see Gentoo bug report #194076)
134 - added autoconf checks for MS Windows
135
136 * general changes:
137 - bugfix: the thread used by an editor plugin didn't die when the
138 editor closed
139 - bugfix: Ringbuffer.h: fill_write_space_with_null() did not zero
140 out all the space. operator--() did not apply size_mask after
141 decrementing the read_ptr. DEFAULT_WRAP_ELEMENTS set to 0 in
142 order to avoid problems with the _NonVolatileReader functions.
143 - bugfix: Stream.h: added a missing divide by BytesPerSample in
144 GetWriteSpace(). Since this function is currently only used in
145 the stream's qsort() compare function, it didn't trigger any
146 bugs.
147 - Resampler.h, Synthesizer.h: cubic interpolation now works in
148 24bit mode too. Faster method to read 24bit words on little
149 endian machines (x86): replaced 3 byte reads + shifts with a 1
150 unaligned 32bit read and shift
151 - experimental support for MS Windows (MIDI input via MME, AUDIO
152 output via ASIO)
153 - made handling of SIGINT signal (Ctrl-C) a bit more robust
154 - support for monitoring the total number of active disk streams
155 (new LSCP commands: GET TOTAL_STREAM_COUNT,
156 SUBSCRIBE TOTAL_STREAM_COUNT, UNSUBSCRIBE TOTAL_STREAM_COUNT)
157
158 * AUDIO driver:
159 - added Windows ASIO low latency audio driver
160
161 * MIDI driver:
162 - added MME Windows MIDI driver
163
164 * LSCP server:
165 - added support for Windows style path / filenames, however with
166 forward slash path separators instead of backslash
167 (i.e. "C:/foo/bar.gig")
168 - allow naughty liblscp to send non-string device parameters within
169 apostrophes as well
170 - added new LSCP commands: "GET FILE INSTRUMENTS <file>",
171 "LIST FILE INSTRUMENTS <file>" and
172 "GET FILE INSTRUMENT INFO <file> <index>" for retrieving informations
173 about an arbitrary instrument file on the system where the sampler is
174 running on
175
176 Version 0.5.0 (15 October 2007)
177
178 * packaging changes:
179 - config.h is not going to be installed along with liblinuxsampler's
180 API header files anymore
181 - only the API relevant header (and source) files will be exported to
182 the Doxygen API documentation (explicitly listed in Doxyfile.in)
183 - added completely new XCode project files for Mac OSX which is now
184 capable to execute our autoconf environment, thus no need anymore
185 to maintain the compile time configuration file (osx/version.h) for
186 OSX manually (patch by Toshi Nagata)
187 - fixed buggy boolean --enable-foo / --disable-foo configure script
188 parameters
189 - global.h now only covers global definitions that are needed for the
190 C++ API header files, all implementation internal global definitions
191 are now in global_private.h
192 - atomic.h is not exposed to the C++ API anymore
193 - no need to include config.h anymore for using LS's API header files
194 - fixed warnings in API doc generation
195
196 * general changes:
197 - replaced the old, confusing MIDI program change mechanism by a
198 flexible MIDI instrument mapper which allows to map arbitrary
199 (MIDI bank, MIDI program) pairs with arbitrary
200 (engine type, instrument file, file index) triplets which will be
201 loaded on the respective channel when such MIDI program change
202 messages arrive, beside that, each entry allows to define a life-time
203 strategy for the instrument, so the user can define whether the
204 instrument should i.e. be permanently loaded (to be able to switch
205 quickly among sounds) or i.e. loaded just on demand when the
206 respective program change arrives (to safe RAM space), as well as a
207 global volume factor for each entry, so the user can adjust the volume
208 dynamic of his mapped instrument collection without having to modify
209 the instrument files, also one can manage arbitrary amount of such
210 MIDI instrument maps and assign each sampler channel individually
211 a certain map, so that i.e. separation of normal instruments and
212 drumkits is possible
213 - new notification events for tracking changes to audio/MIDI devices,
214 MIDI instrument maps, MIDI instruments, FX sends, global volume.
215 - sampler was limited to load max. 200 instruments in the background
216 due to a constant size RingBuffer FIFO which is now replaced by a
217 dynamic (unlimited) size std::list FIFO
218 - added FX sends, these allow to route audio signals to arbitrary audio
219 output channels for being processed by external effect processors
220 (i.e. jack-rack), the send levels are controllable via arbitrary MIDI
221 controllers
222 - global (sampler wide) volume can now be controlled at runtime
223 - Implemented new, improved notification system
224 - fixed compilation errors regarding OSX
225 (patch by Toshi Nagata)
226 - implemented instruments database
227 - added support for escape sequences to the instruments database
228 - added highly experimental support for on-the-fly instrument editing
229 within the sampler's process (by using instrument editor plugins),
230 you'll notice the new "Registered instrument editors:" message on
231 startup, added a new LSCP command:
232 "EDIT CHANNEL INSTRUMENT <sampler-channel>"
233 to spawn a matching instrument editor for the instrument on the
234 given sampler channel, the plugin path can be overridden at compile
235 time with ./configure --enable-plugin-dir=/some/dir
236 - added experimental code for synchronizing instrument editors hosted
237 in the sampler's process to safely edit instruments while playing
238 without a crash (hopefully) by either suspending single regions
239 wherever possible or if unavoidable whole engine(s)
240 - fixed several issues in fundamental "Thread" class: set scheduling
241 policy and priority on thread level, set a minimum stack size for
242 thread (TODO: a reasonable value yet to be tested), bugfix: non-RT
243 threads simply inherited properties of starting thread instead of
244 setting their own policy and priority
245 - minor fix in our automatic stack trace mechanism on crashes, the
246 main process did not wait for the stack trace process to finish
247 its output
248 - fixed some minor memory leaks
249 - reenabled assembly features support, at the moment only for
250 enabling a fast denormal FPU mode (x86 platforms supporting SSE2)
251 - minor assembly fix in x86 features detection (don't use the PIC
252 register, to avoid relocations in the text segment at runtime)
253 - POSIX callback functions of Thread.h are hidden
254
255 * MIDI driver:
256 - dispatch bank select (MSB and LSB) messages
257
258 * audio driver:
259 - the ALSA audio output driver parameters now reflect the correct
260 parameter value ranges for the respective selected sound card
261 (patch by Till Wimmer, a bit fixed and extended)
262
263 * Gigasampler format engine:
264 - if a filter is used and EG2 finishes before EG1, let the voice
265 die when EG2 has finished (this fixes a problem with clicks and
266 voice starvation for some gigs)
267 - playback is no longer disabled during instrument loading
268 - all notes playing on a channel that changes its instrument keep
269 playing with the old instrument until they get a note off
270 command
271 - EG fix: a release value of zero could cause noises or crash
272 - handle MIDI coarse tuning messages (MIDI RPN #0 MSB #2 LSB)
273 - EG fine tuning: when attack is zero the EG starts at a level
274 above max sustain level, which means that there is a decay phase
275 even if sustain is 100%
276 - more EG fixes: the level could sometimes go below zero and cause
277 noises or crashes
278 - minor fix of EGDecay (patch by Toshi Nagata)
279 - fixed compiler error when --enable-override-filter-type was
280 supplied to the configure script (fixes #46)
281 - disk thread: queue sizes are now proportional to CONFIG_MAX_STREAMS
282 instead of a fix value
283 - behavior fix: on MIDI CC# 65 (portamento on / off), 126 (mono mode),
284 127 (solo mode) only kill voices if the respective mode really
285 changed
286
287 * LSCP server:
288 - fixed compile time error for old Bison versions
289 (i.e. v1.28 found on MacOS 10.4, patch by Toshi Nagata)
290 - parser now supports extended ASCII character set
291 (up to ASCII code 255, i.e. includes now umlauts and accents)
292 - filename arguments in LSCP commands now allow to use escape
293 sequences, that is directly literal as one of: \', \", \\, \n, \r,
294 \f, \t, \v, or as octal ASCII code value like \132, or as hex ASCII
295 code value like \xf2) (fixes bug #24)
296 - the following LSCP commands now also support escape sequences for at
297 least one of their text-based parameters (i.e. name, description):
298 "ADD MIDI_INSTRUMENT_MAP", "MAP MIDI_INSTRUMENT",
299 "SET MIDI_INSTRUMENT_MAP NAME", "SET FX_SEND NAME", "CREATE FX_SEND",
300 "SET DB_INSTRUMENT_DIRECTORY NAME",
301 "SET DB_INSTRUMENT_DIRECTORY DESCRIPTION", "SET DB_INSTRUMENT NAME",
302 "SET DB_INSTRUMENT DESCRIPTION", "FIND DB_INSTRUMENTS",
303 "FIND DB_INSTRUMENT_DIRECTORIES"
304 - returns verbose syntax errors (line and column where syntax error
305 occured, the unexpected character and the actually expected, possible
306 character(s), the latter only if less than 5 possibilities)
307 - made sure that LSCP syntax is not affected by gigedit locale
308 settings
309 - bugfix regarding strings parameter lists: all comma separated lists
310 of strings were treated as being one string containing commas
311 (fixes #57)
312
313 * Bug fixes:
314 - fixed crash occurring on certain LSCP scripts (Bug 39)
315 - another thread safety fix for lscp "load engine" and "set
316 channel audio output device"
317 - fixed a crash which occurs when reassigning the same engine
318 on a sampler channel with connected MIDI device
319 - fixed a crash which occurs when changing the number of ports of a MIDI
320 device connected to a sampler channel to number less then or equal
321 to the index of the port to which the sampler channel is connected.
322 - The previous bindings were not been disconnected when altering
323 the ALSA_SEQ_BINDINGS parameter. Introduced a NONE keyword for
324 unsubscribing from all bindings (e.g. ALSA_SEQ_BINDINGS=NONE).
325 - The active stream/voice count statistic was incorrect.
326 - notification events were not been sent for some sampler
327 channel changes
328 - added default min and max values to restrict the number of allowed
329 audio output channels and MIDI input ports
330 - the connection to the PCM interface is now closed when destroying
331 an audio output device
332 - files with slash in their path or filename could not be loaded
333
334 * test cases:
335 - updated and fixed (haven't been touched in a while)
336
337 Version 0.4.0 (24 November 2006)
338
339 * packaging changes:
340 - changed deprecated copyright attribute to license; added ldconfig
341 to post-(un)install steps; added devel package for liblinuxsampler;
342 to linuxsampler.spec (RPM)
343 - install necessary development header files for allowing 3rd party
344 applications to link against liblinuxsampler
345 - liblinuxsampler's API documentation can be generated with 'make docs'
346 (Doxygen required)
347 - added benchmark to automatically detect the best triangle LFO
348 implementation (currently either an integer math solution or a
349 di-harmonic approximation), automatic detection can be overriden
350 with --enable-signed-triang-algo=x and --enable-unsigned-triang-algo=x
351 configure script argument though (mandatory for cross-compilation)
352 - do not automatically pick optimized gcc flags if the user already
353 provided some on his own (as CXXFLAGS)
354 - added compile time option to disable processing of All-Notes-Off MIDI
355 messages
356 - added compile time options to allow disabling the various audio and
357 MIDI drivers
358 - fixed automatic GCC CPU switch detection on PPC
359 (patch by Ebrahim Mayat)
360
361 * Gigasampler format engine:
362 - extensive synthesis optimization
363 (reimplementation of EGs and LFO(s), removed synthesis parameter
364 prerendering and the synthesis parameter matrix in general, splitting
365 each audio fragment into subfragments now where each subfragment uses
366 constant pitch and filter coefficients. The volume coefficient is
367 linearly interpolated inside a subfragment, unless
368 --disable-interpolate-volume is set.)
369 - fine tuning of the EG modulation parameters
370 - improved filter cutoff calculation by adding support for the
371 following gig parameters: Cutoff freq (used when no cutoff
372 controller is defined), Control invert, Minimum cutoff, Velocity
373 curve and Velocity range. The keyboard tracking now scales
374 cutoff frequency instead of resonance.
375 - added support for gig parameter Resonance.
376 - fixed bug in sysex handling (patch by Juan Linietsky)
377 - added global volume attenuation of -9 dB (0.35f) to prevent clipping
378 which can be overridden with --enable-global-attenuation
379 - EG fixes: made the length of "attack hold" stage more
380 accurate. Release stage can now start before attack stage
381 ends. Cancel release didn't work when sustain was zero. Attack
382 time now has a minimal value to prevent clicks.
383 - fixed pitch changes larger than one octave
384 - fixed EG3 (pitch envelope) synthesis which was neutral all the time
385 - implemented portamento mode and solo mode (a.k.a 'mono mode'):
386 all modes can be altered via standard GM messages, that is CC5 for
387 altering portamento time, CC65 for enabling / disabling portamento
388 mode, CC126 for enabling solo mode and CC127 for disabling solo mode
389 - fine tuning of the curves for volume (CC7), pan (CC10 and gig
390 parameter) and crossfade
391 - added support for the "attenuation controller threshold" gig
392 parameter
393 - added smoothing of volume changes caused by control change
394 messages
395 - sample loop parameters are now taken from the DimensionRegion
396 instead of the wave chunk
397 - fixed keyswitching for v3 gigs with a number of keyswitch splits
398 not equal to a power of two
399 - reimplementation of the filter algorithm. The new filter is more
400 accurate and supports all gig filter types, including bandreject
401 and lowpass turbo.
402 - real support for 24 bit samples - samples are not truncated to
403 16 bits anymore
404 - support for aftertouch (channel pressure, not polyphonic
405 aftertouch)
406 - LFO1 behaviour fixed (dampening from max volume instead of
407 amplifying from 0)
408
409 * LSCP server:
410 - fixed application exit on broken pipe error (fixes bug #20)
411 - fixed the notification messages delay due to lack of
412 network activity (fixes bug #26)
413 - fixed parser bug which occured on space(s) within device parameters
414 (e.g. "SET AUDIO_OUTPUT_CHANNEL_PARAMETER 0 0 SOMEPARAM='foo bar'")
415
416 * audio driver:
417 - added aRts audio output driver (by no means RT safe)
418
419 * MIDI driver:
420 - fixed legacy sysex code which caused dispatching of MIDI SysEx
421 messages several times instead of once
422 - API extension for MIDI drivers which already supply exact time stamps
423 for events (i.e. for offline rendering based MIDI drivers)
424 - added checks for bad MIDI data
425
426 * linuxsampler application:
427 - show available sampler engine types on startup
428
429 * general changes:
430 - support for muting sampler channels and solo mode of the same, two new
431 LSCP commands ("SET CHANNEL MUTE" and "SET CHANNEL SOLO") and two new
432 fields ("MUTE" and "SOLO") for command "GET CHANNEL INFO" were
433 introduced for this, the behavior is the same like on a mixer console
434 (patch by Grigor Iliev, a bit adjusted). Also added configure option
435 --enable-process-muted-channels which can be used to enable the
436 processing of muted channels.
437 - support for sostenuto pedal
438 - support for monitoring the total number of active voices
439 (new LSCP commands: GET TOTAL_VOICE_COUNT, GET TOTAL_VOICE_COUNT_MAX,
440 SUBSCRIBE TOTAL_VOICE_COUNT, UNSUBSCRIBE TOTAL_VOICE_COUNT)
441 - fixed some memory management errors
442 - fixed some concurrency problems that could lead to crashes when
443 LSCP commands were executed
444 - fixed crash when instrument loading failed previously on the same
445 sampler channel (fixes bug #36)
446
447 Version 0.3.3 (15 July 2005)
448
449 * packaging changes:
450 - fixed compilation with gcc 4.0
451
452 * Gigasampler format engine:
453 - LFO preprocessing was performed even though the respective LFO was
454 disabled by the instrument patch (minor efficiency fix)
455 - if period time of chosen audio device is too small (< MIN_RELEASE_TIME)
456 for volume ramp downs in the same fragment (needed for current voice
457 stealing implementation) then simply inform the user with a warning
458 about possible click sounds and reduce the volume ramp down
459 appropriately instead of cancelling the audio device connection
460
461 Version 0.3.2 (24 June 2005)
462
463 * packaging changes:
464 - updated autotools build files to compile on OS X
465 (conditional compilation of CoreMIDI and MidiShare drivers)
466 - hand-crafted assembly optimization code can be disabled with
467 './configure --disable-asm' (definitely not recommended)
468 - fixed 'make dist' rule to include all necessary files
469 - require automake (>= 1.5) for 'make -f Makefile.cvs'
470 (due to 'dist-bzip2' automake option)
471
472 * Gigasampler format engine:
473 - support for the gig parameters for "release velocity response" curves
474 - fine tuning of the EGADSR envelope
475 - volume of release triggered samples now depends on note-on velocity,
476 note length and gig parameter "release trigger decay" instead of
477 note-off velocity.
478 - revised voice stealing
479 (fixes crash and endless loop caused by voice stealing)
480 - don't reset scale tuning on instrument or audio output device change
481 - handle key group conflicts right at the beginning of each fragment
482 instead of when the respective voice is actually launched
483 (fixes undefined behavior if stolen voices belonged to a key group -
484 this case was followed by a "killed voice survived" error message)
485 - fixed minor issue with null/silence samples
486 (those stole voices even though they don't need a voice at all which
487 resulted in "voice stealing didn't work out" messages)
488 - don't reset volume, pan, pitch and MIDI controller values on
489 instrument or audio output device change
490
491 * LSCP server:
492 - fixed some crashes (patch by Grigor Iliev, fixes #19)
493 - fixed LSCP event "CHANNEL_INFO" notification
494 (e.g. did not notify on volume changes or MIDI program change events)
495
496 * linuxsampler application:
497 - added command line parameters --lscp-addr and --lscp-port to override
498 default IP address and port of LSCP server
499
500 Version 0.3.1 (24 May 2005)
501
502 * initial release

  ViewVC Help
Powered by ViewVC